Dear Maintainer,
When I run a QEMU image on Debian Bookworm, I observed the following
warning during early boot:
[ 0.000000] NUMA: Warning: invalid memblk node 16 [mem
0x0000000004000000-0x0000000007ffffff]
[ 0.000000] NUMA: Faking a node at [mem
0x0000000004000000-0x00000000bfffffff]
[ 0.000000] NUMA: NODE_DATA [mem 0xbfbf9f80-0xbfbfcfff]
Steps to reproduce:
# on the host:
wget
https://downloads.elxr.dev/elxr-minimal-ostree-imx-12.9.0.0-arm64.img.zst
zstd -d elxr-minimal-ostree-imx-12.9.0.0-arm64.img.zst
cp -f /usr/share/AAVMF/AAVMF_VARS.fd ./
cp -f /usr/share/AAVMF/AAVMF_CODE.fd ./
qemu-img create -f raw boot-image-qemu-single.hddimg 10G
dd if=elxr-minimal-ostree-imx-12.9.0.0-arm64.img
of=boot-image-qemu-single.hddimg conv=notrunc
qemu-system-aarch64 -m 2048 -cpu cortex-a57 -M virt -nographic \
-smp 2,sockets=2,cores=1,threads=1 \
-m 2G,slots=2,maxmem=8G \
-object memory-backend-ram,size=1024M,id=mem0 \
-object memory-backend-ram,size=1024M,id=mem1 \
-numa node,nodeid=0,cpus=0,memdev=mem0 \
-numa node,nodeid=1,cpus=1,memdev=mem1 \
-drive if=pflash,format=raw,file=AAVMF_CODE.fd \
-drive if=pflash,format=raw,file=AAVMF_VARS.fd \
-drive if=none,file=boot-image-qemu-single.hddimg,id=hd,format=raw
-device virtio-blk-device,drive=hd
# on the guest:
root@localhost:~#
root@localhost:~# dmesg | grep -w invalid
[ 0.000000] NUMA: Warning: invalid memblk node 16 [mem
0x0000000004000000-0x0000000007ffffff]
root@localhost:~#
After investigating, I found that the following upstream commit resolves
the issue:
https://salsa.debian.org/qemu-team/edk2/-/commit/e5ec3ba409b5baa9cf429cc25fdf3c8d1b8dcef0
I have backported this patch for bookworm and attached it here for your
consideration.